Replace previous patch adding paths and SerDes modes with patch series
pending upstream adding dedicated drivers for XFI T-PHY and USXGMII PCS,
extends LynxI PCS to be a standalone platform driver and as a consequence
makes much less changes to the actual Ethernet driver mtk_eth_soc.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Let's pick a bunch of useful phylink changes which allow us to keep
drivers in sync with mainline Linux.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
This Qualcomm Gigabit phy driver was mistakenly added because
MT76x8 does not support external phy, and it only supports max
100M full duplex speed.
Fixes: cadf517107 ("ralink: add support for mt7628")
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Backport QCA807x PHY patches merged upstream that introduce the new
concept of PHY package.
Also add in generic config the new Kconfig CONFIG_QCA807X_PHY.
All affected patch automatically refreshed with make
target/linux/refresh.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
u64_stats_init() has been unable to handle NULL pointer since
6.1 kernel. This patch fixes kernel oops on mt76x8 and rt305x
sub-target.
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
In the 5.15 kernel, we use the staging driver version instead of the
downstream file.
Fixes: 88d982e3bd ("ramips: 6.1: mt7621-dma: add hsdma driver to files")
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Fixes errors in the form of:
make[9]: Entering directory '/home/nick/openwrt/build_dir/target-mipsel_24kc_musl/linux-ramips_mt7620/linux-6.1.77'
CC drivers/net/ethernet/ralink/mtk_eth_soc.o
drivers/net/ethernet/ralink/mtk_eth_soc.c: In function 'fe_init':
drivers/net/ethernet/ralink/mtk_eth_soc.c:1368:51: warning: passing argument 2 of 'of_get_mac_address' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
1368 | of_get_mac_address(priv->dev->of_node, dev->dev_addr);
| ~~~^~~~~~~~~~
In file included from drivers/net/ethernet/ralink/mtk_eth_soc.c:26:
./include/linux/of_net.h:16:59: note: expected 'u8 *' {aka 'unsigned char *'} but argument is of type 'const unsigned char *'
16 | extern int of_get_mac_address(struct device_node *np, u8 *mac);
| ~~~~^~~
drivers/net/ethernet/ralink/mtk_eth_soc.c: In function 'fe_probe':
drivers/net/ethernet/ralink/mtk_eth_soc.c:1641:9: error: too many arguments to function 'netif_napi_add'
1641 | netif_napi_add(netdev, &priv->rx_napi, fe_poll, napi_weight);
| ^~~~~~~~~~~~~~
In file included from ./include/linux/etherdevice.h:21,
from drivers/net/ethernet/ralink/mtk_eth_soc.c:21:
./include/linux/netdevice.h:2611:1: note: declared here
2611 | netif_napi_add(struct net_device *dev, struct napi_struct *napi,
| ^~~~~~~~~~~~~~
Signed-off-by: Mieczyslaw Nalewaj <namiltd@yahoo.com>
[split commit and rewrite commit message]
Signed-off-by: Nick Hainke <vincent@systemli.org>
Upstream commit f48dc6b96649 ("spi: Retire legacy GPIO handling") [0]
removed support using GPIOs as chip select. Fix it by replacing cs_gpio
with cs_gpiod.
[0] - f48dc6b966
Signed-off-by: Mieczyslaw Nalewaj <namiltd@yahoo.com>
[split commit and rewrite commit message]
Signed-off-by: Nick Hainke <vincent@systemli.org>
Change fe_hw_set_macaddr and the set_mac parameter to const to fix
errors in the form of:
drivers/net/ethernet/ralink/mtk_eth_soc.c: In function 'fe_set_mac_address':
drivers/net/ethernet/ralink/mtk_eth_soc.c:174:53: error: passing argument 2 of 'priv->soc->set_mac' discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers]
174 | priv->soc->set_mac(priv, dev->dev_addr);
| ~~~^~~~~~~~~~
drivers/net/ethernet/ralink/mtk_eth_soc.c:174:53: note: expected 'unsigned char *' but argument is of type 'const unsigned char *'
drivers/net/ethernet/ralink/mtk_eth_soc.c: In function 'fe_hw_init':
drivers/net/ethernet/ralink/mtk_eth_soc.c:1220:45: error: passing argument 2 of 'priv->soc->set_mac' discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers]
1220 | priv->soc->set_mac(priv, dev->dev_addr);
| ~~~^~~~~~~~~~
drivers/net/ethernet/ralink/mtk_eth_soc.c:1220:45: note: expected 'unsigned char *' but argument is of type 'const unsigned char *'
drivers/net/ethernet/ralink/mtk_eth_soc.c:1222:44: error: passing argument 2 of 'fe_hw_set_macaddr' discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers]
1222 | fe_hw_set_macaddr(priv, dev->dev_addr);
| ~~~^~~~~~~~~~
drivers/net/ethernet/ralink/mtk_eth_soc.c:155:75: note: expected 'unsigned char *' but argument is of type 'const unsigned char *'
155 | static inline void fe_hw_set_macaddr(struct fe_priv *priv, unsigned char *mac)
| ~~~~~~~~~~~~~~~^~~
cc1: all warnings being treated as errors
Signed-off-by: Nick Hainke <vincent@systemli.org>
Upstream dropped support for legacy driver [0]. Rewrite the driver like
the renesas pwm driver [1].
Fixes erros in the form of:
make: *** [/__w/openwrt/openwrt/openwrt/include/toplevel.mk:232: target/compile] Error 1
====== Make errors from logs/target/linux/compile.txt ======
| ^~~~~~~~~~~~~~
drivers/pwm/pwm-mediatek-ramips.c:107:19: note: (near initialization for 'mtk_pwm_ops.free')
drivers/pwm/pwm-mediatek-ramips.c:108:10: error: 'const struct pwm_ops' has no member named 'disable'
108 | .disable = mtk_pwm_disable,
| ^~~~~~~
drivers/pwm/pwm-mediatek-ramips.c:108:20: error: initialization of 'int (*)(struct pwm_chip *, struct pwm_device *, struct pwm_capture *, long unsigned int)' from incompatible pointer type 'void (*)(struct pwm_chip *, struct pwm_device *)' [-Werror=incompatible-pointer-types]
108 | .disable = mtk_pwm_disable,
| ^~~~~~~~~~~~~~~
drivers/pwm/pwm-mediatek-ramips.c:108:20: note: (near initialization for 'mtk_pwm_ops.capture')
cc1: all warnings being treated as errors
[0] - 0829c35dc5
[1] - ec00cd5e63
Signed-off-by: Nick Hainke <vincent@systemli.org>
Upstream dropped slave_id in dai_dma [0]. So drop it also in the mt7620
support patch.
Fixes errors in the form of:
sound/soc/ralink/ralink-i2s.c: In function 'ralink_i2s_init_dma_data':
sound/soc/ralink/ralink-i2s.c:452:17: error: 'struct snd_dmaengine_dai_dma_data' has no member named 'slave_id'
452 | dma_data->slave_id = i2s->txdma_req;
| ^~
sound/soc/ralink/ralink-i2s.c:462:17: error: 'struct snd_dmaengine_dai_dma_data' has no member named 'slave_id'
462 | dma_data->slave_id = i2s->rxdma_req;
| ^~
[0] - https://lore.kernel.org/r/20211122222203.4103644-3-arnd@kernel.org
Signed-off-by: Nick Hainke <vincent@systemli.org>
Upstream changed in ed5c2f5fd10d ("i2c: Make remove callback return void")
the i2c driver's remove function to return no value. Adapt the driver code
to compile with 5.15 and 6.1 like it is done in other projects [0].
Fixes errors in the form of:
make[8]: Leaving directory '/home/nick/openwrt/build_dir/target-mipsel_24kc_musl/linux-ramips_mt7621/linux-6.1.29'
CC drivers/pinctrl/pinctrl-aw9523.o
drivers/pinctrl/pinctrl-aw9523.c:1117:19: error: initialization of 'void (*)(struct i2c_client *)' from incompatible pointer type 'int (*)(struct i2c_client *)' [-Werror=incompatible-pointer-types]
1117 | .remove = aw9523_remove,
| ^~~~~~~~~~~~~
drivers/pinctrl/pinctrl-aw9523.c:1117:19: note: (near initialization for 'aw9523_driver.remove')
cc1: all warnings being treated as errors
[0] - https://gitlab.com/ddcci-driver-linux/ddcci-driver-linux/-/merge_requests/10/diffs
Signed-off-by: Nick Hainke <vincent@systemli.org>
Apply the "109-drivers-mt7621-dma-handle-error-from-device_reset.patch"
directly on the downstream maintained dma driver.
Signed-off-by: Nick Hainke <vincent@systemli.org>
Commit 87dd67f496f7 ("staging: mt7621-dma: remove driver from tree")
removed the mt7621-dma driver. Maintain the driver downstream in the
folder of the other mediatek drivers.
Signed-off-by: Nick Hainke <vincent@systemli.org>
Fix DTS error in LED color/function conversion due to a bug in the
conversion script.
Fixes: 19c45b95db ("ramips: convert to new LED color/function format where possible")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Initial conversion to new LED color/function format
and drop label format where possible. The same label
is composed at runtime.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Drop redundant label with new LED color/function format declared.
This was needed previously when the new format wasn't supported by
leds.sh functions script. Now that is supported this property
can be removed in favor of the new format.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Some of devices in this target have only 8 MiB space and are closing to
borders of usable space. Particularly, TP-Link RE305 v1 already suffers
from this issue[1], where with current partition layout, on release
images, there's not enough space for overlay. So activate small_flash
feature, which will remove some userspace hardening but will gain almost
1 MiB additional flash memory space. Here is small size comparison of
similar device (RE365 v1) with default config + LuCI:
kernel rootfs sysupgrade
current: 2305728 3635044 5964584
small_flash: 1713571 3320132 5047080
1. https://github.com/openwrt/openwrt/issues/14215
Suggested-by: Sander Vanheule <sander@svanheule.net>
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
-+-------------------------+-
| Model | NIC |
-+-------------------------+-
| All | MT7603 + MT7615 |
-+-------------------------+-
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Some MT7915 calibration data consists of two parts. The first part
"eeprom" size is 0xe00. The second part "precal" size is 0x19c10.
Though some devices may not have precal data, it's better to assume
that precal data exists as no users/developers confirm it. On the
other hand, some devices definitely do not contain precal data
because the EEPROM partition size is smaller than the precal NVMEM
cell size.
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
TP-Link RE365 is a wireless range extender, hardware-wise resembles
RE305 with slight changes regarding buttons and LEDs.
Specification
SoC: MediaTek MT7628AN
RAM: 64 MiB DDR2
Flash: 8 MiB SPI NOR
WiFi: 2.4 GHz 2T2R integrated
5 GHz 2T2R MediaTek MT7612EN conncted to PCIe lanes
Ethernet: 1x 10/100 Mbps integrated
LEDs: 6x GPIO controlled
Buttons: 4x GPIO controlled
UART: row of 4 holes marked on PCB as J1, starting count from white
triangle
1. VCC (3.3V), 2. GND, 3. RX, 4. TX
baud: 57600, parity: none, flow control: none
Installation
1. Open web management interface.
2. Go to Settings > System Tools > Firmware upgrade.
3. Select "Browse" and select the OpenWrt image with factory.bin suffix.
4. After selecting "Upgrade" firmware writing process will start.
5. Wait till device reboots, power LED should stay solid when it's fully
booted, then it's ready for configuration through LAN port.
Additional information
With how device manufacturer patrtitioned the flash memory, it's possible
that with default packages set, initial factory.bin image won't be
created. In such case, try to reduce packages amount or use older release
for initial conversion to OpenWrt. Later You can use sysupgrade.bin
image with full set of packages because OpenWrt uses unpartitioned flash
memory space unused by vendor firmware.
Reverting to vendor firmware involves converting firmware using
tplink-safeloader with -z option (can be found in ImageBuilder or SDK)
and forcibly applying converted firmware as sysupgrade.
Known issues
WARNING: after removing casing of the device one is exposed to high
voltage and is in a risk of being electrocuted.
Caution when interfacing whith bootloader, saving its environment either
by issuing "saveenv" or selecting option "1: Load system code to SDRAM
via TFTP." in boot menu, any of those will lead to overwriting part of
kernel. This will lead to need of firmware recovery. The cause of this
issue is bootloader having environment offset on flash at 0x40000,
while kernel starts from 0x20000.
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
[Wrap long line in DTS]
Signed-off-by: Sander Vanheule <sander@svanheule.net>
This commit:
1. Removes deprecated "label" property from the dts leds subnnodes;
2. Updates buttons and leds dts description according to kernel docs
examples.
Scope: devices well known to me.
Run-tested: TP-Link ec330-g5u, WiFire S1500.nbn
Signed-off-by: Mikhail Zhilkin <csharper2005@gmail.com>
MT7688 devices use the "mt7628an.dtsi" as the template. And RT3052
devices use the "rt3050.dtsi" as template. Therefore, we need to add
the corresponding system controller compatible strings to make them
work properly.
Fixes: 1f818b09f8 ("ramips: add proper system clock and reset driver support for legacy SoCs")
Fixes: #14305
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Use reset controller to reset mt7620 ethernet phy instead of directly
writing system control registers. The reset line of "ephy" is 24, so
the DTS resets properties have been updated to get the correct reset
signal.
Tested on HiWiFi HC5861.
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Use reset controller to reset mt7620 frame engine instead of directly
writing system control registers.
Tested on HiWiFi HC5861.
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Flash: 16MB SPI NOR flash (Macronix MX25L12805D)
Based on the manufactured datasheet this chip is capable of 50MHz.
We dont enable fast-read as mt7621 are only capable of 44mhz in a read state.
Tested on this unit without any issues.
Signed-off-by: David Bentham <db260179@gmail.com>
The rt305x series SOC have two UART devices,
and the one at bus address 0x500 is disabled by default.
Some boards do not even have a pinout for the first one,
so use the same one that the kernel uses at 0xc00 instead.
This allows the lzma-loader printing to be visible
alongside the kernel log in the same console.
Tested-by: Lech Perczak <lech.perczak@gmail.com> # zte,mf283plus
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Before this was reworked, in the file for mt7621 subtarget
(target/linux/ramips/image/lzma-loader/src/board-mt7621.c)
the "Transmitter shift register empty" bit TEMT was used instead of
the "Transmitter holding register empty" bit THRE,
but after the rework, this value was labeled as the THRE bit instead.
Functionally there is no difference, but this is confusing to read,
as it suggests that the subtargets have different bits for the same
register in UART when in reality they are exactly the same.
One can use either bit, or both, at user's descretion
in order to determine whether the UART TX buffer is ready.
The generic kernel early-printk uses both,
(arch/mips/kernel/early_printk_8250.c)
while the ralink-specific early-printk uses only THRE,
(arch/mips/ralink/early_printk.c).
Define both bits and rewrite macros for readability,
keep the same values, as changing which to use should be tested first.
Ref: c31319b66 ("ramips: lzma-loader: Refactor loader")
Signed-off-by: Michael Pratt <mcpratt@pm.me>
The native bus address for UART was entered for rt305x UART_BASE,
but the bootloaders have memory space remapped with the same
virtual memory map the kernel uses for program addressing at boot time.
In UBoot, the remapped address is often defined as TEXT_BASE.
In the kernel, for rt305x this remapped address is RT305X_SYSC_BASE.
(arch/mips/include/asm/mach-ralink/rt305x.h)
Because the ralink I/O busses begin at a low address of 0x10000000,
they are remapped using KSEG0 or KSEG1, which for all 32-bit MIPS SOCs
(arch/mips/include/asm/addrspace.h)
are offsets of 0x80000000 and 0xa0000000 respectively.
This is consistent with the other UART_BASE macros here
and with MIPS memory map documentation.
Before the recent rework of the lzma-loader for ramips,
the original board-$(PLATFORM).c files also did not
use KSEG1ADDR for UART_BASE despite being defined,
which made this mistake easier to occur.
Fix this by defining KSEG1ADDR again and actually use it.
Copy and paste from the kernel's macros for consistency.
Link: https://training.mips.com/basic_mips/PDF/Memory_Map.pdf
Fixes: c31319b66 ("ramips: lzma-loader: Refactor loader")
Reported-by: Lech Perczak <lech.perczak@gmail.com>
Signed-off-by: Michael Pratt <mcpratt@pm.me>
The ESW core needs to be reset together with FE core, so after the
relevant reset controller lines are moved under FE, drop rst_esw and all
related code, which would not execute anyway, because rst_esw would be
NULL. While at that, ensure that if reset line for EPHY cannot be
claimed, a proper error message is reported.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Co-developed-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
Signed-off-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
[Split out of the bigger commit, provide commit mesage, refactor error
handling]
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Failing to do so will cause the DMA engine to not initialize properly
and fail to forward packets between them, and in some cases will cause
spurious transmission with size exceeding allowed packet size, causing a
kernel panic.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Signed-off-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
[Provide commit description, split into logical changes]
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Failing to do so will cause the DMA engine to not initialize properly
and fail to forward packets between them, and in some cases will cause
spurious transmission with size exceeding allowed packet size, causing a
kernel panic.
This is behaviour of downstream driver as well, however I
haven't observed bug reports about this SoC in the wild, so this
commit's purpose is to align this chip with all other SoC's - MT7620
were already using this arrangement.
Fixes: #9284
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Failing to do so will cause the DMA engine to not initialize properly
and fail to forward packets between them, and in some cases will cause
spurious transmission with size exceeding allowed packet size, causing a
kernel panic.
This is behaviour of downstream driver as well, however I
haven't observed bug reports about this SoC in the wild, so this
commit's purpose is to align this chip with all other SoC's - MT7620
were already using this arrangement.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Failing to do so will cause the DMA engine to not initialize properly
and fail to forward packets between them, and in some cases will cause
spurious transmission with size exceeding allowed packet size, causing a
kernel panic.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Signed-off-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
[Provide commit description, split into logical changes]
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Enabling the FE core too early causes the system to hang during boot
uncondtionally, after the reset is released. Increate it to 1-1.2ms
range.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Signed-off-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
[Split previous commit, provide rationale]
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Use devm_reset_control_array_get_exclusive to register multiple
reset lines in FE driver. This is required to reattach ESW reset to FE
driver again, based on device tree bindings.
While at that, remove unused fe_priv.rst_ppe field, and add error
message if getting the reset fails.
Fixes: 60fadae62b ("ramips: ethernet: ralink: move reset of the esw into the esw instead of fe")
Co-developed-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
Signed-off-by: Maxim Anisimov <maxim.anisimov.ua@gmail.com>
[Split out of the bigger commit, provide commit mesage, refactor error
handling]
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
